Landscapes

Use tupperware and single-well plates for experiments which require high reproducibility and low cost, use 3D-printing for unique designs and more complex networks.

3D-Printing

3D-printing is overall relatively easy to perform in the lab, however it can be time consuming and there are limits to the size you are able to print. One advantage of the printing system we have is its high resolution, but this also means that you are restricted to the size of approximately 150 x 150 x 150mm. I would recommend using this system to print individual units and then build a larger system together (like legos). This will allow for the creation of flexible, complex systems that best utilize the properties of the printer. Also make sure to 3D-print compatible lids so that the collembola can’t jump out!

  • Software
    • Use OnShape to create designs. Students can make free accounts and it’s compatible with the printer.
    • Use PreForm to upload designs to make adjustments before printing and to select correct resin, resolution, etc.
      • I have had issues with wireless connection between the printer and the device with PreForm, so instead I have used the USB cord to connect my laptop to the computer to upload the design.
    • Pipeline: design in OnShape, download design as .stl file, open in PreForm, adjust as necessary, upload print. Save print jobs as .form files.
  • Printer
    • Model is Form 2.
    • I’ve only ever used Color Base resin with Cyan from previous user (almost empty). We also have a full Clear resin cartridge and lots of other color pigments.
    • If the printer screen is frozen (unable to select a print), try unplugging and replugging the printer.
    • Wear gloves when working with resin!
  • Wash
    • Model is Form Wash (1st Generation).
    • The user before me did not properly clean the resin from the wash, meaning that the fan at the bottom is stuck in place. You might be able to fix this by dissolving the hardened resin in IPA, but it isn’t fully necessary for the print to get washed.
    • I used IPA as the solvent.
  • Cure
    • Model is Form Cure (1st Generation).
    • Curing is not necessary for all prints, but a good idea to optimize the properties of the resin.

Tupperware

  • Mainly used for maintenance populations.
  • Ensure that a corner of the tupperware is not fully sealed to allow for airflow. Unless the populations are very dense the collembola will not climb the walls/lids of the tupperware. Lightly tapping the lids of the containers before opening them will reduce the likelihood of them jumping/spilling out.
  • If you observe mold growing on the lids, remove it with tweezers or a paper towel.

Single-Well Plates

  • Link to order.
  • Dimensions (128 x 86mm) work well for imaging with the D7000.
  • Since they are more shallow than the tupperware containers, when lifting up the lids be careful of collembola jumping out. Slowly lifting the lid straight up will reduce this.
Example Single-Well Plate. Note the collembola naturally tend towards the edges of the plate. Also note that the substrate is properly saturated given its dark color.
Example Single-Well Plate. Note the collembola naturally tend towards the edges of the plate. Also note that the substrate is properly saturated given its dark color.

Substrate Preperation

Substrate is a mixture of Plaster of Paris (PoP), active charcoal, and water.

  • Use 4:20 ratio of charcoal:PoP to create master mix.
  • For single-well plates, use 24g of master mix. Then add 10mL of water to each plate and mix thoroughly (especially the corners!) before tapping on the bench to ensure that the substrate will be level and smooth.
    • Optionally: after the plaster is set, use black ink marker to blackout the walls/edges of the plate to reduce glare during imaging.
  • For maintenance tupperware, add ~1cm of master mix (in height).
  • For 3D-prints/more complex structures, make sure that there is at least ~1 cm of space/wall above the height of the substrate to ensure that the collembola won’t jump out.
    • Filling these structures with substrate can be a bit challenging; use any tools that might be helpful! I used pipette tips to ensure the substrate was able to reach all edges/corners.
  • Leave the substrate/plate overnight before adding water or collembola to allow for the plaster to fully set. Before adding collembola, re-hydrate the plaster with a few mLs of water to ensure optimal conditions for the collembola.

Collembola Care

Collembola (Folsomia candida, aka springtails) are generally super easy to care for! They only require a few things, including airflow, water, and food:

Airflow

  • Don’t use an airtight container to store the collembola!
  • Using 3D-Printed structures or Single-Well Plates, there should be enough airflow naturally that you can fully close the lids.
  • Using tupperware, make sure to leave one corner of the tupperware unsealed so that the lid still covers the container and the collembola can’t jump out, but also the container is not fully sealed.

Water

  • Optimal conditions for the collembola includes moisture! This means that the substrate should always be saturated. The easiest way to determine this is if the charcoal looks fairly dark and not ashy. There should not be puddles and all of the water should be absorbed by the substrate.
  • For maintenance populations, use a squeeze bottle (not precise).
  • For experiments, use a pipette. For Single-Well Plates
  • For maintenance populations I would recommend watering the populations every 3-4 days. However they are resilient and can absolutely survive for ~1-2 weeks without water if need be.

Food

  • Use Fleischmann’s Quick-Rise yeast. Link to order.
  • For maintenance populations, use the lid of the yeast bottle or a metal spatula to portion out food. Try to scatter the yeast out across the container so that there are no large clumps/concentrated areas. For a healthy population (100+ individuals), start with approximately a dime sized quantity of yeast and see how long it takes for them to eat all of it. Ideally, you should need to feed the populations every 3-4 days to ensure that mold does not grow from uneaten old food.
  • For experiments, use tweezers to count exactly how many yeast particles you are adding (tedious but precise!). Also note location where you are adding the yeast (center of plate, corners, etc.) as this will affect the movement of the collembola.

Transferring Collembola

  • Use the manual aspirator (test tube with two plastic tubes attached to the lid) to move individuals.
    • This can be stressful to the individuals so try to work quickly and as gently as possible!
    • Plastic tube with green tape = in contact with the collembola.
    • Plastic tube with no tape = in contact with your mouth.
    • Unscrew the lid of the test tube to empty the container of any collembola you aspirated.
  • If you need to transfer a large number of individuals, you can carefully tilt and tap the old container over a new container and let the collembola sort of fall into the new container gently.

Other

  • Use the manual counter to count individuals in each experimental plate to track population dynamics over time.
  • You can use a light system on a timer for consistency with experiments. Make sure to run experiments in W4/10K to eliminate inconsistent effects of natural light!
  • Mold can sometimes grow in the containers, especially in the maintenance containers where food is not added as precisely. I recommend trying to remove the mold whenever possible without disturbing the individual collembola, using tweezers or a paper towel. If the mold issue gets out of hand and seems to be continuously growing, I would recommend transferring the surviving population into a new container.

Camera & Imaging

  • Use the Nikon D7000 camera for all imaging purposes.
    • Refer to D7000 Manual.
    • Settings I used:
      • Mode = Manual (M)
      • Release-Mode = Single frame (S)
      • Shutter speed = 160
      • Aperture = F3.2
      • White balance = Preset manual (PRE) set to an image I took under d-0
      • Image quality = JPEG fine (FINE)
      • Image size = Large (L)
  • We have 3 different lenses in the lab:
  • Use the K&F Concept tripod in the lab to situate the camera properly. Make sure everything is fastened tight on the tripod before imaging so that all of the images are consistent in their framing!
    • You can try to keep the camera held in the tripod throughout a multi-day experiment to make sure the framing stays consistent. After several days, you will have to remove the battery and charge it but this can be done even if the camera is still held by the tripod. Same goes for removing the SD card for analysis.
  • I found it helpful to change the Storage folder (under SHOOTING MENU) for each plate/day to make it easier to keep track of the data. For example, plate 1 images would be located in the folder 100, plate 2 images would be located in the folder 101, etc.
  • Use Interval timing shooting (under SHOOTING MENU) to automatically take a certain number of photos at a specified interval. Image sequences created using Interval timing shooting are smaller and easier to work with than videos.
    • I used the interval of 1 second, 600 intervals, and 1 shot per interval = 1 image per second for 10 minutes.
      • I thought this interval/number of shots worked well to encompass both the quick, stochastic movement of individual collembola as well as the slower overall movement of the whole population. But of course this setting should be used differently for different experiments.

Image Analysis

  • I used Fiji for all of my image analysis. I was mainly looking for individual tracks and movement data for individual collembola.
  • Image analysis steps:
    • Open the folder containing the image sequence that you want to analyze into Fiji.
      • For an image sequence of 600 photos, this should take ~7.5 minutes.
    • Convert the RGB image into 8-bit: Image > Type > 8-bit.
      • For an image sequence of 600 photos, this should take ~2.5 minutes.
    • Crop the images so that the edges are the same as the edges of the internal part of the single-well plate to avoid the analysis tool detecting false positives which are reflections of collembola on the walls: Image < Crop (after selecting a rectangular region to crop).
      • For an image sequence of 600 photos, this should take ~1 minute.
    • Open the Fiji plugin called TrackMate: Plugins > Tracking > TrackMate. Make sure the Z variable is set 0 to 0, and T is set to 0 to the number of images taken - 1 (eg. 599).
      • Select LoG detector.
      • Set the Estimated object diameter to ~56 pixels. Play around with this using the Preview button to see if this seems like an accurate size to detect the collembola.
      • Set the Quality threshold to ~0.116. Again use Preview to see if this looks correct (minimizing false negatives and positives) across different time points.
      • Check Sub-pixel localization.
      • The subsequent Detection takes ~4.5 minutes for an image sequence of 600 photos.
      • You can skip through Initial thresholding.
      • Select Simple LAP tracker.
      • Set the Linking max distance to ~112.0 pixels. I set this to the length of 2x collembola based on observations I made about their movement, but you might want to adjust this.
      • Set the Gap-closing max distance to ~672.0 pixels. I set this to the length of 12x collembola, but you might want to adjust this. I set it to this because assuming the linking max distance is 112.0 pixels, and the gap-closing max frame gap is 6, then 112.0x6=672.0.
      • Set the Gap-closing max frame gap to ~6. Again, I made this decision based on observations I made about their movement, but you might want to adjust this.
      • The TrackScheme, Tracks, and Spots buttons will have tons of movement data that can be further analyzed (eg. each track’s duration, displacement, mean speed).
      • If you keep pressing the Next button, you can perform a Capture overlay which will produce a video of the tracks that you can save as a .AVI file. You can play around with the Display options (eg. make the spots the same color as their corresponding Track index).
  • You can find some examples of raw image sequences and the corresponding tracks of the collembola produced using TrackMate on the SD card in the lab.